i

TITLE:

ANALYSIS OF SPATIAL AND TEMPORAL VARIABILITY OF RAINFALL IN SINGIDA

REGION TANZANIA FROM THE YEAR 2001 TO 2010.

RESEARCH PROJECT II

BY

MUSIBA, MUSIBA MASAMBA.

SUPERVISOR: EMMANUEL L. NDETTO (Mr.)

A SPECIAL PROJECT REPORT SUBMITTED IN PARTIAL FULFILMENT OF THE

REQUIREMENTS OF THE DEGREE FOR THE BACHELOR OF SCIENCE IN

ENVIRONMENTAL SCIENCES AND MANAGEMENT OF SOKOINE UNIVERSITY OF

AGRICULTURE, MOROGORO TANZANIA

JULY 2011

ii

Abstract

The present study examined the spatial and temporal variability of the annual rainfall observations

over the Region of Singida in Tanzania for the period 2001 - 2010 using 7 rainfall stations available in

the region. Rainfall variability was characterized by using the Coefficient of variability and other

statistical measures. In order to come up with critical variability the region was sub-dived into three

sub-regions according to the monthly and annual rainfall distribution. The relationship between the

rainfall amount and variation was also investigated. The results indicated that a sub-region which

receives relatively high annual rainfall also seems to experience highest rainfall variability. The

northern part of the region received highest mean annual rainfall and has highest rainfall variability of

689.5mm and 30% respectively. The overall region average annual rainfalls is associated with the

pattern that average annual rainfall decreases southwards and the temporal variation of the rainfall

shows a statistically significant decreasing trend in the last decade.

KEY WORDS: Singida; spatial variability; temporal variability; rainfall, rainfall variability, rainfall

trend.

iii

Acknowledgement

It is an honor for me to acknowledge and heartily thankful to my supervisor Mr. Emmanuel L. Ndetto

of Faculty of Science, Sokoine University of Agriculture whose inspirational instructions, guidance,

support and encouragement from the initial to the final level enabled me to develop an understanding

of the subject. I am sure it would have not been possible without his help and assistance in doing this

wonderful special project on the topic ANALYSIS OF SPATIAL AND TEMPORAL

VARIABILITY OF RAINFALL IN SINGIDA REGION. This also helped me in reading a lot of

Research reports and I came to know about so many new things. I am really thankful to him and God

bless him abundantly.

Secondly I would also like to thank the administration of Tanzania Meteorological Agency (TMA) for

providing the rainfall data for Singida which was used in this project.

Lastly, I offer my regards and blessings to my friends, classmates and my dear sister Mrs. Happy

Masamba Musiba whose support both materially and spiritually helped me in respect to the

completion of this special project.

Above all I thank my Almighty GOD who is my first priority in all angles of my life in this world. By

his GRACE I live.

THANKS AGAIN TO ALL WHO HELPED ME.

iv

Dedication

This work is dedicated to my mother, Veronica Weseja Kitundu, without her caring supports it would

not have been possible, and to the memory of my father, John Masamba Musiba, who passed away, on

a love of reading and respect for education. Their support and encouragement are forces towards

fulfillment of my degree program at Sokoine University of Agriculture.

v

Declaration

I Musiba, Musiba Masamba of registration number ESM/D/08/T.9851, hereby declare to the Senate

of the Sokoine University of Agriculture that I am the author of this knowledge and the work

presented in my own and has not been submitted for a Bachelor Degree requirement at any other

University.

Signature ………………….

Date ………………………..

vi

Copyright

No one should reproduce, store in any reproducible system, or transmit in any form of or by means of

electronics, photocopy or doing anything to any part of this Special Research Report, without the prior

written permission of the Author or Sokoine University of Agriculture on my behalf.

vii

Table of Contents

Abstract .................................................................................................................................................... ii

Acknowledgement .................................................................................................................................. iii

Dedication ............................................................................................................................................... iv

Declaration ............................................................................................................................................... v

Copyright ................................................................................................................................................ vi

Table of Contents ................................................................................................................................... vii

List of figures and Tables ....................................................................................................................... ix

Abbreviations and acronym ..................................................................................................................... x

CHAPTER ONE ...................................................................................................................................... 1

Introduction .............................................................................................................................................. 1

1.1 Background information ................................................................................................................ 1

1.2 Problem statement .......................................................................................................................... 3

1.3 Problem justification and significance ........................................................................................... 4

1.4 Objectives of the study ................................................................................................................... 6

CHAPTER TWO ..................................................................................................................................... 7

Literature review ...................................................................................................................................... 7

2.1 Climatology .................................................................................................................................... 7

2.2 Weather systems influencing rainfall ........................................................................................... 9

2.3 Rainfall analysis ............................................................................................................................. 9

2.4 Rainfall variability..................................................................................................................... 9

CHAPTER THREE ............................................................................................................................... 13

Methodology .......................................................................................................................................... 13

3.0 Description of study area.............................................................................................................. 13

3.1 Data collection and methods used. ............................................................................................... 14

CHAPTER FOUR .................................................................................................................................. 17

Results and Discussion .......................................................................................................................... 17

4.1 Results .......................................................................................................................................... 17

4.2 Discussion ............................................................................................................................... 24

4.2.1 Rainfall characteristics in the Region .............................................................................. 24

4.2.2 Rainfall trend in the Region ............................................................................................. 24

4.2.3 Rainfall intensity .............................................................................................................. 25

4.2.4 Temporal Rainfall variability ........................................................................................... 25

viii

4.2.5 Spatial Rainfall variability . ............................................................................................. 26

CHAPTER FIVE ................................................................................................................................... 28

Conclusion and Recommendation ......................................................................................................... 28

5.1 Conclusion .................................................................................................................................... 28

5.2 Recommendation .......................................................................................................................... 28

References .............................................................................................................................................. 30

ix

List of figures and Tables

Figure 1: Singida Map showing the sub-regions and position of each station ............................................ 16

Figure 2: Graph representing mean monthly rainfall ................................................................................... 21

Figure 3: Graph showing long term mean monthly rainfall ........................................................................ 21

Figure 4: mean annual rainfall and rainfall trend for Singida ...................................................................... 22

Figure 5: Stations mean annualrainfall……………………………………………….............…….……..22

Table 1a: The Annual average rainfall (AAV), SD and CV of each station ............................................... 17

Table 1b: The Annual average rainfall (AAV), SD and CV of each sub-region ......................................... 18

Table 2: Region annual mean rainfall (mm) ................................................................................................ 18

Table 3, Region monthly mean rainfall (mm) ............................................................................................. 19

Table 4, Stations` mean annual rainfall (mm) ............................................................................................. 19

Table 5: Standard Deviation and Coefficient of variability of monthly rainfall in a decade....................... 20

x

Abbreviations and acronym

1. CV -------- -Coefficient of variability

2. ENSO ----- -El Niño–Southern Oscillation.

3. FAO -------Food and Agriculture Organization

4. IPCC ------- Intergovernmental Panel on Climate Change.

5. ITCZ ------- The Inter tropical Convergence Zone

6. SD --------- Standard Deviation.

7. TMA -------Tanzania Meteorological Agency.

1

CHAPTER ONE

Introduction

1.1 Background information

Intra-annual rainfall variability refers to the distribution of rainfall within a year (Obasi, 2003).

Rainfall is one of the most important meteorological phenomenon’s, regarding to climatic

condition and climate classification. The amount of rainfall is one of the most variable

meteorological data. The variation of the amount of precipitation in arid and semi-arid regions is

higher than in humid one. Consequently, the prediction of precipitation in time and/or in space is

very difficult and uncertain in such regions. However, the precipitation that does occur in these

regions is the most important aspect of the available water resources (Nicholson &

Entekhabi,1997).

According to Nyenzi (1999); Zorita and Tilya (2002) the inter-annual rainfall variations in the

last few decades are the causes of great stress to the farming activities, crop production and crop

yield in various areas in Africa. This variability affects various aspects of plant growth and

yields; consequently, alter crop productivity. From an analysis of the year 2000 on rainfall

conditions in East Africa, FAO (2001) concluded that a long-term change in rainfall has occurred

in the semiarid and arid zones of East Africa and most of its causes are natural. The need for in-

depth study and understanding of spatiotemporal rainfall variability is of utmost importance.

These analyses will allow agricultural producers as well as policy makers to know the impact of

spatial-temporal variation in rainfall on crop yield for better management, productivity and

profitability.

2

Changes in seasonal patterns of rainfall are also considered to be one of important key

ecological factors (Wimalasuriya et al., 2009). Rainfall is a key factor in shaping the vegetation,

hydrology, and water quality throughout the earth. As rainfall critical to agriculture its temporal

and spatial variability is also important in determining other hydrological functions. Along with

temperature, the occurrence and variability of rainfall, to a large extent determine which crops

can be grown in different regions throughout the world. As additional insight into precision

farming is gained, the importance of rainfall variability becomes more apparent. An accurate

understanding of precipitation characteristics is critical to optimizing farm production and to

precision farming (Bosch et al., 1998).

Rainfall is also one of the main factors affecting the sustainability of irrigation areas in terms of

dictating the need for irrigation to grow crops and determining the drainage requirements needed

to avoid shallow water table conditions and secondary soil compaction. The variability of rainfall

in the upper catchments of rivers determines the water available in storage reservoirs that can be

used for irrigation to meet crop water use requirements (Khan et al., 2001).

Singida region is in semi-arid tropics characterized by low and highly variable rainfall in both

space and time in such a way to limit potential crop yields in the areas. This high degree of

rainfall variability, when combined with relatively low income of urban and rural inhabitants,

could restrict the management and overall agriculture productivity around this region. As

smallholder farmers in Singida region Tanzania practice rain fed agriculture, they are therefore at

high risk of crop failure given the unpredictable nature of the rains as well as the rainfall

variability. The situation reduces assurance of household food security (Ngana, 1991).

Understanding how variable the rainfall is could help in studies intending to know how the

3

variable rainfall and farmer’s available resources all together can affect crop productivity.

Wealthy farmers may be better able to cope with this rainfall variability. However poorer

farmers suffer the most where the rainfall is low and variable as they have little time to plant as

they plant late due to lack of financial resources. The study can design planting options for both

the rich and poor farmers that enable them to cushion the impact of poor and variable rainfall

(Kavishe, 1993).

1.2 Problem statement

The economy of Tanzania depends significantly on rain fed agriculture, which is highly

vulnerable to the amount and distribution of rainfall. Both spatial rainfall anomalies over the

region and temporal variability within the wet season itself can be prominent and lead to

significant deficiencies. In some areas of Tanzania, agriculture is limited by the length of the rain

season while in others it is amount limited (Mhita, 1984). Associated with global warming, there

are strong indications that rainfall changes are already taking place on both the global and

regional scales. Future climate changes may involve modifications in climatic variability as well

as changes in averages. Thus, rather than mean values, inter annual variability is defined by

many as one of the most important indicators of the reliability of the rainfall resource (Bosch et

al., 1998). The implications of these changes are particularly significant for areas like Singida in

the central Tanzania which are already under stress. The region suffer a water shortage through a

combination of a dry climate (or a highly seasonal rainfall regime) and excessive demand

(Kavishe, 1993). The region is thus a transitional climate zones where it can be hypothesized that

rainfall variability may have the greatest effects. Therefore, understanding the character of the

spatial and temporal variability of precipitation is critical (Mosaedi and Kavakebi, 2010).

4

1.3 Problem justification and significance

There are challenges which are still numerous in Tanzania due to strong spatial and temporal

variability nature of rainfall (Ogallo, 1989; Nyenzi et al., 1999; Zorita and Tilya, 2002). Since

the past many years, this has been an evident challenge of rainfall patterns in central zone of

Tanzania, Singida in particular, which had been referred to as trend for decreasing precipitation

and changes in seasonal distribution (Murray et al., 2008). In some region, a significant part of

precipitation variability occurs at the micro-scale and spatial coherence which is quite small.

Under these conditions, the significance and magnitude of changes at a regional scale are

difficult to ascertain. Nevertheless, it is important to assess the changes that may be occurring on

a spatial and a temporal basis. In this research, a complete description of intraregional rainfall

variability and changes is of great interest, especially in these areas with strongly contrasting

rainfall regimes.

The role of rainfall as a resource in crop production has been an area of interest for many

researchers studying the major droughts regions in Tanzania. The region of Singida being part of

the central Tanzania has dry weather due to scanty rainfall in the area. Some studies like that

done by Lema (1998) revealed a general outlook that central Tanzania rainfall is extremely

variable on all temporal scales. But the study did not look an insight on spatial scales.

Furthermore, past studies were done while Singida had no synoptic meteorological station which

started operating officially in 2000. Either the temporal scales studied were too general by not

considering specific parts of the region. Therefore there is a need for other studies to be done in

the region to provide an insight of the direction and the significance of rainfall trends at local

scales. These studies of rainfall variability focusing on large scale areas, for example, the area

5

average rainfall series of the whole country or meteorological subdivisions could be of less

significance at local scales. Dewar and Wallis (2008) also commented on such general studies

that would be of no use for local agriculture, particularly in places where rainfall is very high or

very low, especially for crops.

The area of interest is semi-arid and the rainfall in most of recent seasons is below normal.

Therefore studying Singida region is in need since the effects of rainfall scarcity and drought are

dramatic and obvious while the effects of the oversupply of rainfall are much less noticeable

except in extreme cases which often go unnoticed for many years.

As mentioned earlier, agriculture in the study area is largely rain fed. Therefore, the amount of

rain that falls in the area is critical to agricultural planning and development. Rainfall studies in

Singida in the past years could have met great obstacles because there has been no any synoptic

Meteorological station since then until May 2000. In this case only few (probably no) studies

have been developed to use the data obtained from this station for 10 years since it become into

operation. The station helps to give a good picture of rainfall because the observations are

assumed be of good quality rather than those from volunteer stations. Either most studies are

developed country wise and not in a local scale to the extent of a region. This paper analyses

rainfall variability on local scales important to the social-economic benefit of farmers, water

managers, energy suppliers and public policy makers in the regional and Tanzania in general. In

developing a better understanding of regional forcing of drought and flood years, reliable long-

range forecast models may be developed.

6

1.4 Objectives of the study

This special project use the existing rainfall data to create a synthesis paper generally analyzing

the rainfall variability in Singida region with the following specific objectives:-

a) Analyzing the spatial distribution of rainfall. This reveals how rainfall differs from one or

more stations location to another.

b) Analyses to show temporal variation of rainfall.

The above objectives answers the following questions,

a) Which areas experience the most or least variability of annual rainfall amount and which

areas the variability is prominent within the period of study?

b) The average trend of rainfall in the last ten years, does the general trend increasing or

decreasing or constant?

c) Which areas within the region have the highest or lowest amounts of rainfall?

7

CHAPTER TWO

Literature review

2.1 Climatology

According to Emma (2005), Tanzania is not homogeneous from a climatic point of view. Some areas

have bi-modal rains i.e. have two distinct rainfall seasons comprised of the long rains (“Masika”)

between March-May and short rains (“Vuli”) between October-December. This pattern of rainfall is

typical of north- eastern, north-western (Lake Victoria basin) and the northern parts of the coastal belt.

The rest part experiences uni-modal rains which usually start in November and ending in April.

Tanzania being located very near to the equator, most of her places experience equatorial weather and

climate, with hot days and rainy monsoons. The proximity of Tanzania to the equator could have made

her weather not to undergo vast seasonal variations in temperature. The climate could remain favorable

between June and September. Since the country has varied geographical landscape features, the

patterns of weather keep on changing on moving from one place to another. The weather varies on the

mountainous, the coastal plains and in the northern, eastern and southern part of the country. The

weather in the mountainous areas often reaches below the freezing point and remains cool all

throughout the year. Along the coasts the weather is tropical and carries lots of moisture from the

Indian Ocean, making the places humid.

The central, western and northern Tanzania mainly comprises of highlands and plateaus. Though the

temperature remains moderate during most of the time in the year, it often soars high in the summer

months. Humidity in these places is relatively low compared to other parts of Tanzania.

8

Weather in the central part of Tanzania is dry due to scanty rainfall in this area. The areas are semi-

arid and the rainfall is below normal. The northeast and southwest regions in Tanzania are primarily

mountainous areas, where weather conditions are pleasant during daytime and summer, and is

utterly chilly in winter (Mary & Majule, 2009).

According to Paavola (2003), Tanzania experiences periodic droughts and flooding. Rainfall

increases from normal in the first year of a two year El Nino episode and decreases from normal in

the second year of the cycle. However, rainfall patterns in Tanzania follow the opposite cycle: first

year of El Nino brings less than normal amount of rain and the second year brings more of it than

usual. El Nino influences the length of rainy season and short fall rains in the equatorial region

which experiences two rainy seasons in a year. La Nina reduces rainfall all over Africa during its

first year and increases it in the second year. La Nina influences the amount of rainfall especially

during the primary rainy season. However, their effect on Tanzania is less pronounced than those of

El Nino. Finally, variations in rainfall are strongly related to sea surface temperature (SST)

variations in the Indian Ocean and the Atlantic which may sometimes alter standard oscillation

outcomes.

Variability of rainfall is predicted to increase as the result of increased frequency of El Nino events

(IPCC, 2001). Changes in the mean temperature and rainfall, and the increased variability of

rainfall, are together use to prolong the length of dry seasons annually and to increase the severity

of periodic droughts. This is usually pronounced in the interior part of the country which

experiences reduced rainfall.

9

2.2 Weather systems influencing rainfall

In the tropics seasonal precipitation patterns are strongly influenced by seasonal changes in

quasi-stationary pressure systems, regional convergence zones, and monsoonal circulations.

Other factors known to influence precipitation over East Africa include tropical storms, easterly

waves, jet streams, the continental low-level trough, and extra-tropical weather systems (Ogallo,

1989).

2.3 Rainfall analysis

Since precipitation is unevenly distributed in space and time, its analysis presents challenges.

Precipitation can have properties, such as precipitation rate, amount, duration, spatial

distribution, temporal distribution, and the organization of precipitation systems. A statistical

study can be performed for the spatial and temporal distribution of rainfall at annual and seasonal

time-scales (Longley, 2008).

2.4 Rainfall variability

Precipitation over an area is usually estimated from a network of rain gauges. Most gauges are

placed near permanent settlements rather than distributed evenly. Thus, gauge-only maps are

missing data over the ocean or inaccessible locations. Types and location of gauges produce

additional sources of uncertainties. Although there is an abundance of precipitation in the tropics

compared with the mid latitudes, most of that precipitation is concentrated in a few locations at

any given time (Schreck & Semazzi, 2007).

10

According to Nieuwolt (2008), annual rainfall has greater geographical variability than

temperature and a wide disparity exists between the amounts received in the humid and dry

tropics. The origins of these variations are uncertain, although there are many studies that try to

connect them with climate forcing factors such as solar activity, oceanic and atmospheric

circulation indices such as the El Niño–Southern Oscillation (ENSO). Other major influences on

the distribution of annual precipitation are water vapor content, large-scale mechanism for lifting

air to its condensation level, relief, amount and type of cloud condensation nuclei. The highest

average annual precipitation amounts are on the windward side of tropical mountains, whereas

the lowest are found far inland, in the subtropics on the western coasts where high pressure

dominates, and on the leeward side of mountains. Annual rainfall amounts provide only a partial

picture of tropical rainfall.

Globally, precipitation frequency varies greatly from continent to continent and across some

regions. A distinctive pattern of highly variable, moderately variable and least variable regions is

evident. When precipitation occurs is as important as the amount of precipitation. Studies in

Tropical Meteorology are interested in understanding the tropical seasonal precipitation cycle

and its variability because tropical precipitation is a critical part of the global water and energy

cycle. The seasonal distribution of precipitation influences how people live. Bearing in mind that

agriculture is a dominant part of many economies in the tropics and agricultural activities are

controlled by the start, duration, and end of the rainy season (Xie & Arkin, 1997).

Lui´s et al., (2000) used the time series of seasonal rainfall (winter, spring, summer and autumn)

to test for trends in mean values and changes in inter annual seasonal rainfall in the region of

Valencia which is characterized by highly irregular distribution and a high inter annual

variability of seasonal rainfall in the four seasons. The most relevant result that was found was

11

the great spatial variability of the seasonal rainfall trend in the area, where positive, negative and

zero trends occurred in different parts of the region. As a consequence, it was concluded that

seasonal rainfall trends seemed to be strongly affected by local factors in a manner not

previously known. There seemed to be a double gradient affecting trends of seasonal amounts

and seasonal rainfall variability. The latitudinal gradient, north–south, seemed to divide the area

according to rainfall generation processes while the longitudinal gradient, coast–inland, seemed

to describe more adequately the spatial distribution of variability trends on a seasonal scale, and

could have suggested a temporal change in the frequency of rainfall processes.

Buytaert et al., (2006) studied the spatial and temporal rainfall patterns of the western mountain

range of the Ecuadorian Andes and showed that clear intraday pattern were possible to be

distinguished. Rain gauges at a mutual distance of less than 4000 m were strongly correlated,

with a Pearson correlation coefficient higher than 0.8. Within this perimeter, spatial variability in

average rainfall was very high; however the seasonal variation was low. Significant correlations

were found between average daily rainfall and geographical location, as well as the

topographical parameters slope, aspect and topography.

Richard et al., (2000) used an original dataset for the Dobruja, a low plateau region of

southeastern Romania, to perform a statistical study of the spatial distribution of rainfall at

annual and seasonal time-scales. In spite of the relatively flat topography of the area, annual

rainfall seemed to be characterized by a strong gradient between the coast and the interior. The

analysis revealed that this spatial distribution aroused from a combination of specific seasonal

patterns among which the warm season presented the most acute spatial differentiation. Altitude

and distance from the Black Sea were analyzed as potential factors of this spatial distribution,

reinforced during the warm season. Correlation and stepwise regression indicated that the

12

altitude effect is moderate compared with the distance from the sea. The study generalized that,

causes of strong spatial variations in rainfall and its seasonal progression were due to factors

such as altitude, aspect, site (valley, valley-side, and summit), distance from the sea, latitude, and

so on. Longitude was considered redundant for that present context, because of its close

correlation with distance from the Black Sea. In addition, there was no observed association

between latitude and rainfall amount: there was no meridional precipitation gradient whatever

the season. Unfortunately, it had not been possible to study local site factors, and therefore

considered the local topographical climatology, because of lack of detailed data. On the other

hand, altitude and, especially, distance from the Black Sea appeared at times to affect the spatial

distribution of precipitation and its seasonal progression.

In general, tropical precipitation is highest during the summer and lowest during the winter.

Strong variability in the intensity and distribution of tropical precipitation also occurs at sub-

seasonal time scales. Precipitation is sensitive to small spatial and temporal variations in the

general circulation. Dramatic swings from anomalously low to high precipitation rates can occur

in one year, as occurred between the strong El Niño of 1997-98 and the La Niña of 1998-99. It

should be noted that not all areas of high variability are responses to ENSO (El Niño - Southern

Oscillation). The ENSO phenomena is known to be a fundamental and quasi-periodic feature of

the ocean-atmosphere system, with periodicities ranging from 2 to 7 years (Halpert and

Ropelewski, 1992). Some extreme rainfall anomalies in East Africa have been associated with

ENSO (Ogallo, 1988). The seasonal rainfall patterns in East Africa are controlled by the seasonal

migration of the ITCZ. The complex topographical patterns, the existence of large lakes,

variations in vegetation type and land–ocean contrasts give rise to high spatial and temporal

variation in precipitation over the region (Ogallo et al,. 1998).

13

CHAPTER THREE

Methodology

3.0 Description of study area

Singida is located in Tanzania between Latitudes 4○

04'S - 7°12'S and Longitudes 33°42'E-35°

28' E and it is at an elevation of 1547 meters. The climate of the region is basically of an inland

equatorial type modified by the effects of altitude and distance from the Equator. The region

forms part of the semi-arid central zone of Tanzania experiencing low rainfall and short rainy

seasons which are often erratic with fairly widespread drought of one year in four. Singida region

has a unimodal rainfall regime, which is concentrated in a period of six months from November

to April. The long-term mean annual rainfall is 674 mm with a standard deviation of 179 mm and

a coefficient of variation of 28.7%. The long-term mean number of rainy days is 49 with a

standard deviation of 15 days and a coefficient of variation of 30.6%. Generally rainfall in the

region is low and unreliable.

The annual mean, maximum and minimum monthly mean daily relative humidity is 80.6%,

86.0% (February) and 73.4% (July) respectively (Lema, 2008). The maximum and minimum

monthly mean daily pan evaporation is 6.6 mm/day (November) and 5.2 mm/day (January) with

standard deviations of 1.2 mm/day and 0.8 mm/day respectively (ibid). Temperatures vary

according to altitude. The annual mean, maximum and minimum monthly mean daily

temperatures in the region are 22.0°C, 24.4°C (November) and 19.3°C (July) respectively. The

average annual daily sunshine hours are 7.9 h/day. The maximum and minimum monthly mean

daily sunshine hours are 9.2 h/day (September) and 6.5 h/day (January) respectively. The region

has one synoptic station located within the municipality and a number of rainfall gauges installed

at different places within the region.

14

3.1 Data collection and methods used.

If we intend to draw meaningful conclusions on rainfall trends in space, a dense and

homogeneous network is required. Climate variability studies show that large-scale changes in

annual precipitation over land areas can be documented when a dense network database is used

and is especially relevant in areas with strongly contrasting rainfall regimes like in semi-arid and

arid tropical regions (Vinnikov et al., 1990).

Rainfall data for 14 stations within the regional political boundary were collected from the

archives of the Tanzania Meteorological agency. Rainfall recording started in different years, but

for purposes of this study the data used for the analysis was starting from 2001 through 2010.

The stations are namely; Ihanja, Ilongero, Kilimatinde, Kinampanda, Kiomboi Administration

Office, Manyoni Bomani, Msingi Training Centre, Ndago, Ndunguti, Singida Synoptic station,

Singida Bomani , Senkenke Hydromet Station, Shelui and Puma station. But most stations have

missing data in such a way that only seven stations were qualified to be used for analysis which

are Ilongero, Kiomboi, Manyoni, Msingi Training centre, Nduguti, Singida Synoptic station and

Singida Bomani station. The following were the criteria used for sorting out the stations and

years for analysis within 2001 and 2010.

The station with rainfall data for more than five (5) years.

The years with consecutively missing data less than three.

By the use Microsoft excel software the following were calculated.

Annual total and average for each station in every year.

Every month average for the entire period for each station.

15

The average total annual rainfall for each station.

Total annual average for all stations in each year.

The mean, standard deviation and coefficient of variation of total annual rainfall for all

years for each station.

The mean, standard deviation and coefficient of variation of each year average for the

whole region.

The data consist of monthly totals from seven stations, whose locations are shown in Fig 1

below. For these stations, the average record length is 10 yrs from 2001 to 2010, which is the

period of interest for this study. 14.3% of the observation years are not included in the analysis

because are either missing or do not meet the criteria (i.e. have missing data more than three

consecutively). For the available years, 3.7% of rainfall data were reported as missing (m).

The stations divide the study area into three sub regions namely, the northern part consisting of

Kiomboi, Msingi Training Centre, Ilongero and Nduguti stations, the central part by Singida met

and Singida Bomani, and the southern part by Manyoni station. The sub-regions are shown on

the map Singida below.

16

Shinyanga Region

Lake Eyasi Manyara Region

4◦S

Sekenke Kiomboi Nduguti LEGEND

Msingi T. Centre

Northern part Ilongero

Singida bomani

Singida met. Station

Tabora Region SINGIDA REGION

Central part

6○S

Itigi Manyoni

Dodoma Region

Southern part

Mbeya Region Iringa Region

Figure 1: Singida Map showing the sub-regions and position of each station

Regional boundary

Main road

Railway

Sub-region boundary

8◦S

17

CHAPTER FOUR

Results and Discussion

4.1 Results

(a) The total annual rainfall for the stations has annual average (AAV), standard deviation (SD)

and coefficients of variation (C.V) in table 1a and 1b.

Table 1a: The Annual average rainfall (AAV), SD and CV of each station

Station Ilongero Kiomboi Manyoni Msingi

T.Centre

Nduguti Singida

Met

Singida

Bomani

AVV(mm) 579.18 854.23 643.15 299.00 627.40 654.66 670.38

SD 175.77 153.77 132.14 118.40 210.37 188.79 171.42

CV 0.30 0.18 0.21 0.40 0.34 0.29 0.26

On considering an individual station, results reveal that Nduguti station has a high variability

with highest standard deviation of 210mm per year and rainfall variability at Kiomboi is least

with high average annual rainfall of 854 mm per year as indicated in table 1a. In sub region

summary below it was found that highest rainfall variability is 30mm per year in the Northern

Singida in which Nduguti station is located.

18

Table 1b: The Annual average rainfall (AAV), SD and CV of each sub-region

Sub-region Northern part Central part Southern part

AVV(mm) 689.521 662.11 643.15

CV% 30.5 27 21

In the table 1b above the region is divided into sub-regions, the northern part has the highest

average annual amount of rainfall followed by the central part and the southern part having the

least amount. The rainfall variability follows the same trend; it decreases southward. The sub-

region with high average annual rainfall (northern part, 689.5mm) also has highest rainfall

variability (30%).

(b) The average annual rainfall for the region in each year.

Table 2: Region annual mean rainfall (mm)

YEAR 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

AVERAGE 689.5 840.4 456.4 644.0 409.4 697.8 582.7 617.3 886.4 460.1

CV 25.4%

Throughout the ten years of this study, the region received highest average rainfall in 2002

(840.44mm) and the lowest amount in 2005 (409.37mm). The rainfall variability is by 25.4%

between years.

19

c) Total monthly average rainfall in the region.

Table 3, Region monthly mean rainfall (mm)

Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Total 7685.6 5308.4 6855.4 3258.4 1012.5 128.6 78.7 10.1 203.6 511.6 3293.1 7331.1

Average 134.84 93.13 120.27 57.16 17.76 2.26 1.38 0.18 3.57 8.98 57.77 128.62

The above table 3, indicate that for the last decade the region received the highest average

monthly rainfall in the month of January followed by December and March. The driest months

start in May through October.

(C) Each station average total annual rainfall.

Table 4, Stations` mean annual rainfall (mm)

Year /

Station

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

Ilongero

681.9

883.9

392.3

645.7

432.7

640.1

596.8

360

Kiomboi

898.8

1152.8

662.1

789.5

722.3

775.1

886.2

947

Manyoni

584.8

733.6

703.3

739.5

349.9

642.3

640.3

751.5

Msingi

438.8

181.8

347.8

157.2

434.7

203.9

328.7

Nduguti

779.7

961.3

461.8

558.8

335.6

727.2

567.4

Sgd Met

603.2

872

394.7

644.1

417.9

880.5

691.4

692.4

890.3

460.10

Sgd Bom

588.7

840.7

398.6

782.8

450

861.9

603.7

685

822

Average

689.52

840.44

456.37

644.03

409.37

697.78

582.66

617.30

886.43

460.10

20

d) Mean monthly variability.

Table 5: Standard Deviation and Coefficient of variability of monthly rainfall in a decade

Monthly Average

Month

/Year

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

2001 267.2 81.0 129.5 67.2 4.2 1.1 10.2 0.0 3.8 16.6 33.5 75.3

2002 209.9 90.1 177.5 57.7 50.9 0.0 0.0 0.1 0.4 3.7 133.5 183.6

2003 96.8 43.9 82.7 48.0 59.7 3.7 0.1 0.2 1.8 23.3 2.9 139.1

2004 92.9 167.6 146.5 87.4 4.1 0.0 0.0 0.0 22.9 1.1 51.6 129.0

2005 108.8 75.6 138.9 63.2 12.5 0.0 0.0 0.0 0.0 8.7 18.5 25.2

2006 90.3 77.2 87.4 73.0 9.5 4.6 0.0 1.3 0.0 0.0 110.6 296.5

2007 215.3 130.6 76.3 38.5 7.1 9.0 1.1 0.3 0.0 5.3 9.6 152.9

2008 80.7 144.9 176.4 44.5 0.3 0.4 1.8 0.0 2.9 26.9 102.3 93.9

2009 172.8 63.0 188.6 77.2 26.1 7.0 0.0 0.0 0.0 4.1 120.5 227.2

2010 94.3 130.7 64 5.8 3.7 0 0 0 0 0 29.6 132

SD 54.41 42.06 49.20 24.48 21.80 3.48 0.64 0.42 7.51 10.06 52.16 77.52

CV 2.63 2.39 2.58 2.30 0.82 0.74 2.06 0.46 0.42 0.89 1.17 1.88

CV% 263 239 258 230 822 74 206 46 42 89 117 188

Considering table 5 above, the monthly average rainfall is variable between 42% and 822%. The

variability is random between wet and dry months. Both wet and dry months have either high or

low rainfall variability.

21

(e) Graph of average total monthly rainfall for the period of study.

Figure 2: Graph representing mean monthly rainfall

The mean monthly rainfall in the above figure 2 depicts that the wettest months are December,

January and March. The driest months are May, June, July, August, September and October.

(f) Graph of study area long-term average monthly rainfall (TGDD, 2002)

Figure 3: Graph showing long term mean monthly rainfall

135

93

120

57

18

02 01 0004

09

58

129

0

20

40

60

80

100

120

140

160

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecMonth

Mon

thly

mea

n R

ain

fall

(m

m)

22

The figure 3 above shows the character of long term monthly rainfall average. The wettest months

are December, January, February and March while the driest months are May, June, July, August,

September and October.

(g) Graph of average annual rainfall and the trend line

Figure 4: mean annual rainfall and rainfall trend for Singida

In figure 4 above, the years with highest average rainfall were 2002, and 2009 while 2001, 2004,

2006 received slightly higher monthly rainfall than the long term average monthly rainfall. The

driest years were 2003, 2005 and 2010. The trend shows that region has negative rainfall trend

for the past ten years.

0

100

200

300

400

500

600

700

800

900

1000

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010Year

Aver

age

an

nu

al

rain

fall

(mm

) Trendline

23

(h) Graph of station’s average annual rainfall.

Figure 5: Stations mean annual rainfall

Considering figure 5 above, Kiomboi rainfall station recorded highest annual rainfall while

Msingi Training centre recorded little amount compared to other stations.

0

100

200

300

400

500

600

700

800

900

Ilongero Kiomboi

office

Manyoni

Bomani

Msingi

centre

Ndunguti singida met

station

Singida

Bomani

Station

Aver

age

an

nu

al

rain

fall

(mm

)

24

4.2 Discussion

4.2.1 Rainfall characteristics in the Region

Analysis of mean monthly rainfall of the Region in general, as in figure 2 indicates a significant

monthly variation with one peak starting mostly in November through April. Between 2001 and

2010, the Region experienced a relatively similar pattern of rainfall relative to the long term

pattern (refer to fig 2 and 3). The effective wet months are December, January and March while

in February although still wet, there is a slightly decrease in monthly rainfall. The rainfall

experiences unimodal rainfall pattern as shown in figure 2. The rains start mainly at the month of

November and few cases in October through April and in few years until May. The wettest

months in the region are December and January which receives monthly rainfall between 130mm

and 150mm as illustrated on the two graphs in figure 2 and 3. The months of May and October

receive rainfall less than 20mm they are dry while driest month being June, July, August and

September which may receive none.

4.2.2 Rainfall trend in the Region

Despite of showing similar patterns, the long-term monthly average for the given years (fig 3)

when compared to ten year average (fig 2), it can give by what amount and percentage the

rainfall in the last10 years were either higher than or lower than the long-term average. Rainfall

averages appears to have decreased approximately by 66mm which is about 9.5% lower than the

long-term average. Either in figure 4, it indicates that the region experiences an increasing dry

regime relative to the long term average, as from 2001 through 2010. The ten-year moving

average line point out a relative decrease in average rainfall since the trend line goes down

significantly from each year. The region has had a couple of years with rainfall average of more

25

than 30 per cent less than the average, in 2003, 2005, 2007 and 2010. The highest rainfalls on

record for this region are +42% in 2009 and +34% in 2002. The two lowest annual rainfalls had

been -34% in 2005.

4.2.3 Rainfall intensity

Although the analyses suggest the decrease in annual rainfall in the last decade, this could not

tell whether the intensity decreased or not. The number of rain days could have been decreased

or increased. It was not easy in determining intensity of rainfall whether increased or not since

the rainfall data provided was in terms of monthly totals which can also give annual totals. The

only thing could be done with an annual rainfall is to compute the annual rainfall rate, which

won't be very useful. So for example if 480mm of rain falls on a location....that’s 40mm/month

or 0.056mm/hr. Of course, that's useless information, since it wasn't raining 0.056mm/hr the

entire year, but rather there were periods of little or heavier rain and periods of no rainfall at all.

4.2.4 Temporal Rainfall variability

4.2.4.1 Mean monthly rainfall variability

Table 5 based on data between 2001 and 2010 indicates monthly rainfall variability and trend

with special interest in the months of November, December, January, February, March, April,

May and July. From the table it is apparent that monthly rainfall show considerable variations.

Mean monthly total rainfall for the wet months (e.g. January) ranged between 80.7mm (2008)

and 267.2mm (2001) and from 0 to 22.9mm for driest months (e.g. September). The

corresponding coefficient of variation of mean total monthly rainfall for 10 years, are highest

during the wet months i.e. 117.45%, 187.65%, 262.61%, 238.84% 257.67% and 229.8%

26

respectively in November, December, January, February, March and April than dry months.

Highest Coefficient of variation recorded in wet months suggests planning uncertainties

especially for agriculture and other rainfall dependent activities.

4.2.4.2 Mean annual rainfall variability.

Amount of rainfall per annum is the most used indicator of rainfall pattern among planner (Hella,

J.P 2004). The data collected from seven stations indicate that mean rainfall per annum ranged

between 409.37mm in 2005 to as much as 886.43mm in 2009. The range is high to almost half

the highest rainfall recorded in the region. The maximum average annual rainfall does not show a

great deal of temporal variability. This is due to the fact that about 60% of the stations in the last

decade have their maximum average annual rainfall in the month of January and about 30% in

December which are the wettest months.

The most evident is the variability of annual rainfall between years. Referring to table 2 the

annual average rainfall is variable by 25.4%.

4.2.5 Spatial Rainfall variability .

Taking note on figure 1b, the region when divided into sub-regions show differences in mean

annual rainfall and variability. The northern region shows a relatively higher variability with

coefficient of variability of about 30% with compared to other regions. According to Longley

(2008) the coefficient of variation (CV) is common indicator of the amplitude of variability,

which is defined as the ratio between the standard-deviation and the mean. But the same part

seemed to be wettest in the past 10 years by having the average annual rainfall of 689.5mm

compared to 674mm of the long term average.

27

The central part rainfall is variable by 27% and received an annual average rainfall of 662.11mm

which is slightly below the long-term average. So this part experienced mild condition as

compared to other parts.

The southern part was the driest of all by receiving the average annual rainfall of 643.15mm as

compared to the long term average. The same area rainfall is least variable by 21% when

compared to other parts.

Therefore out of the three sub-regions, the northern part exhibits the most profound rainfall

variability (in 10 years) with more than 30%. The southern part shows the least variability by

21% and the central part by 27% nearly to that of the northern part. Since the northern part

seemed to have received a pronounced high rainfall compared to other sub-regions. The

latitudinal gradient, north–south, divide the region according to rainfall averages. This describes

more adequately the spatial distribution of variability trends on an annual scale. This means the

average annual rainfall received in the region decreases southward. The latitude differences

depict the relation with annual rainfall although this study can go that far as to explain the causes

of spatial variation.

28

CHAPTER FIVE

Conclusion and Recommendation

5.1 Conclusion

Annual rainfall in Singida could be highly variable over the years. Based on the historical data in

general, the given year’s rainfall may be anywhere between nine per cent lower than the long-

term average. This variability of rainfall may be resulting into droughts. Having a dry year or a

“drought” is not an exceptional circumstance. Rainfall seems has decreased in Singida during

last decade as a sudden shift, rather than gradual trend. The ten-year moving average trend-line

of historical annual rainfall provides an insight to the above statement and earlier mentioned

rainfall variability. The ten-year average has been persistently nearly to ten per cent below the

long-term average. This could be the type of situation that Singida is experiencing currently. It is

this type of a multi-year drought which could be seen as exceptional. This may not only be an

agricultural drought, but it can develop into a hydrological drought where stream flows, dam

levels and even groundwater levels can fall. The rainfall variability and amount follows

latitudinal gradient, by decreasing as you move toward the south. The rainfall trend could have

implication on agriculture activities, water resources and other economical activities in the

region.

5.2 Recommendation

As the effects of droughts to human welfare is known. Further research may be needed however,

to determine the most appropriate indicator for defining a drought. First, what’s the most suitable

number of years to calculate the medium-term average of annual rainfall? By what percentage

29

should the multi-year moving average be lower than the long-term average annual rainfall? For

how many years should it stay at that level continuously?

Further research may also be required to examine the direct links between rainfall variability and

agricultural activities. As mentioned before, multi-year droughts could proceed from an

agricultural drought to a hydrological drought. The latter may have significant implications for

irrigation-dependent activities. Understanding how rainfall variability has affected various

economical activities will allow policy-makers to develop policies that support on-farm

adaptation effort.

The future trend should be studied and predicted so that to the Government Authorities should

start to plan alternatives to type of agriculture and other economic activities to be implemented.

This may try to mitigate the effects which might be caused by future trend of decreasing rainfall

if any.

30

References

Bosch, D. D. and Davis, F.M. (1998). Rainfall variability and spatial patterns for the southeast.

Southeast Watershed Research Laboratory USDA-ARS. Tifton, Georgia.

Buytaert, W, Celleri, R, Willems, P, Bie`vre, B, and Wyseure, G. (2006). Spatial and temporal rainfall

variability in mountainous areas: M.A.R.D.I., Serdang, Selangor, Malaysia.

Dewar, R.E, and Wallis, J.R. (2008). Geographical Patterning of Inter annual Rainfall

Variability in the Tropics and Near Tropics: An L-Moments Approach. J Climate 12:3457–

3466.

Emma, L. (2004). Climate Change Perception and Adaptation among Rural populations in The

Great Ruaha Basin. Institute Of Resource Assessment (IRA), University Of Dar Es Salaam

Tanzania

FAO (2001). Climate Variability and Change: A Challenge for Sustainable Agricultural

Production. Committee on Agriculture.

Halpert, M.S. and Ropelewski, C.F. (1992). Surface temperature patterns associated with the

Southern Oscillation‟, J. Climate, and 5,577–593.

Hella, P. J. (2004).Rainfall variability impacts on farmers’ crop management strategies.

International Crops Research Institute for the Semi-Arid Tropics. Bulawayo, Zimbabwe.

IPCC (2001).Climate Change 2001: Synthesis Report

Kavishe, M.M. (1993). Climate change in East Africa... Tanzania Meteorological Society

Report, Dar es Salaam. Tanzania.

31

Khan,S and Short, L. (2001). Assessing the Impact of Rainfall Variability on Water tables in

Irrigation Areas, Griffith Laboratory, Research Station Road .Griffith, NSW, 2680.

Kijazi, A .L and Reason, C.J.C. (2004). Relationships between intra seasonal rainfall variability

of coastal Tanzania and ENSO.

Lema, A (1998). Links between rainfall variability on intraseasonal and interannual scales over

central Tanzania and regional circulation and SST patterns. J Climatol 18: 40–53.

Longley, R.W (2008), Measures of the variability of precipitation. Department of Physical and

Environmental Sciences, University of Toronto, Scarborough, Ontario, Canada.

Lui´S, M. D, Ravento´ S J, Lez-Hidalgo , J.C , Sa´ Nchez, J.R. and Cortina, J. (1990). Spatial

Analysis of Rainfall Trends in the Region of Valencia (East Spain). Department of Ecology

University of Zaragoza, Spain.

Mary, A. L and Majule, A. E. (2009). Impacts of climate change, variability and adaptation strategies on

agriculture in semi arid areas of Tanzania: The case of Manyoni District in Singida Region,Tanzania.

Institute of Resource Assessment (IRA), University of Dar es Salaam. Tanzania.

Mhita, M.S. (1984). The use of water balance models in the optimization of cereal yields in

seasonally-arid tropical regions. PhD Thesis. University of Reading.

Mosaedi, A and Kavakebi, G. (2010). Statistical Characteristics of precipitation Data in Arid

and Semi-arid Regions of Khorasan Razavi Province, Ferdowsi University of Mashhad,

Faculty of Natural Resources and Environment .Iran .

Murray, L. A, Orindi, V.A and Mulenga, H. (2008), Adapting to Climate Change in East Africa:

A strategic approach, Int J Climatol 30:29–56.

Ngana, J. O. (1991). Selection of probability distribution for estimating reliability seasonal

rainfall in Semi-arid Central Tanzania. IRA paper no. 79, UDSM.

32

Nicholson, S.E and Entekhabi D. (1987). Rainfall variability in equatorial and southern Africa:

relationships with sea surface temperatures along the southwestern coast of Africa. J Clim

Appl Meteorol 26:561–578

Nieuwolt , S. (2008). Tropical rainfall variability — The agro climatic impact. Agrometeorological Unit.

M.A.R.D.I., Serdang, Selangor, Malaysia.

Nyenzi, B.S. (1999). Mechanisms of East African rainfall variability: PhD Thesis. – The Florida

state University, College of Arts and Science.

Obasi, G.O.P. (2003b). “Regional conference on climate change and food Sustainability” in the

21st century: Lagos, Nigeria.

Ogallo, L. J. (1989). The spatial and temporal patterns of the East African seasonal rainfall

derived from principal component analysis. International Journal of Climatology, 9: 145–

167.

Ogallo, L.J, Indeje,M and Semazzi, F.H.M.(1998).ENSO signals in East African rainfall

seasons. J Climatology 18:34–43.

Ogallo, L.J. (1988). Relationship between seasonal rainfall in East Africa and Southern

Oscillation. J Climatology 8:34–43.

Paavola, J. (2003). Vulnerability to Climate Change in Tanzania: Sources, Substance and

Solutions.

Richarda, Y, Mihailescu, I and Planchon,O. (2000). Spatial distribution of the precipitation in

Dobruja (Romania:Black Sea). CNRS University of Burgundy, Dijon, France .

Schreck, J, and Semazzi, F. H. M. (2007). Variability of the recent climate of Eastern Africa.

North Carolina State University. Department of Marine, Earth and Atmospheric Sciences.

USA.

33

Tanzania Government Directory Database (2002). Retrieved on 7th

May, 2011 from

www.tanzania.go.tz/govdirf.html.

Vinnikov, K.Y, Groisman, P.Y and Lugina, K.M.(1990). Empirical data on contemporaly

global climate changes (temperature and precipitation). Journal of Climate 3: 662–677.

Wimalasuriya, R, Arthur, H, Tsafack, E and Larson, K. (2009).Rainfall Variability and its

Impact on Dryland Cropping in Victoria. Canberra, Australia.

Xie, P and Arkin, P.A. (1997). Global precipitation: A 17-year monthly analysis based on gauge

observations, satellite estimates and numerical model outputs. Bull Amer Meteor Soc 78:

2539–2558.

Zorita, E and Tilya, F.F. (2002). Rainfall variability in northern Tanzania in the march-may

season (long rains) and its links to large scale climate forcing. Clin. Res. 20: 31-40.

.